With availability of new genotyping technology, in particular SNP arrays as well as the coming next generation sequencing, efforts of mapping genes of human diseases/traits have been focusing on genetic association study. Family-based and population-based studies are two commonly-used designs of genetic association study. In contrast to population-based study, family based study is robust to bias due to population stratification. However, family-based study is often less powerful than population-based study because association information between families is not used due to its susceptibility to the bias of population stratification. The aim of this application is to develop a new analytic tool to fully utilize data as well as maintain the robustness to population stratification. Specifically, we propose a new statistic approach under the Empirical Bayesian framework, in which the bias of population stratification of individual loci is estimated so that the amount of information between families contributed to the testing statistic shrinks based on the bias. To demonstrate the validity and superior performance of the proposed approach compared to approaches available, we plan to evaluate it by extensive simulations and apply it to the family genome-wide association data in dbGaP. PUBLIC HEALTH RELEVANCE: With availability of new genotyping technologies, efforts of mapping genes of human diseases/traits have been focusing on genome-wide association study (GWAS). Family-based and population-based studies are two commonly-used designs used in GWAS, and each of them has unique advantages and disadvantages. The aim of this application is to develop a new analytic tool to make use of advantages of both designs - to maintain the robustness against population stratification and to achieve higher power. The proposed study has the potential to facilitate the research for identifying novel loci related to complex diseases.